arXiv subjects

Feng Chen

Publications and source records attributed to Feng Chen.

At least 289 records · Page 16Linked to original sources

arXiv2021

Local Aggressive Adversarial Attacks on 3D Point Cloud

Deep neural networks are found to be prone to adversarial examples which could deliberately fool the model to make mistakes. Recently, a few of works expand this task from 2D image to 3D point cloud by using global point cloud optimization. However, the perturbations of global point are not effective for misleading the victim model. First, not all points are important in optimization toward misleading. Abundant points account considerable distortion budget but contribute trivially to attack. Second, the multi-label optimization is suboptimal for adversarial attack, since it consumes extra energy in finding multi-label victim model collapse and causes instance transformation to be dissimilar to any particular instance. Third, the independent adversarial and perceptibility losses, caring misclassification and dissimilarity separately, treat the updating of each point equally without a focus. Therefore, once perceptibility loss approaches its budget threshold, all points would be stock in the surface of hypersphere and attack would be locked in local optimality. Therefore, we propose a local aggressive adversarial attacks (L3A) to solve above issues. Technically, we select a bunch of salient points, the high-score subset of point cloud according to gradient, to perturb. Then a flow of aggressive optimization strategies are developed to reinforce the unperceptive generation of adversarial examples toward misleading victim models. Extensive experiments on PointNet, PointNet++ and DGCNN demonstrate the state-of-the-art performance of our method against existing adversarial attack methods.

cs.CV↗

arXiv2021

Boosting Cross-Lingual Transfer via Self-Learning with Uncertainty Estimation

Recent multilingual pre-trained language models have achieved remarkable zero-shot performance, where the model is only finetuned on one source language and directly evaluated on target languages. In this work, we propose a self-learning framework that further utilizes unlabeled data of target languages, combined with uncertainty estimation in the process to select high-quality silver labels. Three different uncertainties are adapted and analyzed specifically for the cross lingual transfer: Language Heteroscedastic/Homoscedastic Uncertainty (LEU/LOU), Evidential Uncertainty (EVI). We evaluate our framework with uncertainties on two cross-lingual tasks including Named Entity Recognition (NER) and Natural Language Inference (NLI) covering 40 languages in total, which outperforms the baselines significantly by 10 F1 on average for NER and 2.5 accuracy score for NLI.

cs.CL↗

arXiv2021

Non-autoregressive Transformer with Unified Bidirectional Decoder for Automatic Speech Recognition

Non-autoregressive (NAR) transformer models have been studied intensively in automatic speech recognition (ASR), and a substantial part of NAR transformer models is to use the casual mask to limit token dependencies. However, the casual mask is designed for the left-to-right decoding process of the non-parallel autoregressive (AR) transformer, which is inappropriate for the parallel NAR transformer since it ignores the right-to-left contexts. Some models are proposed to utilize right-to-left contexts with an extra decoder, but these methods increase the model complexity. To tackle the above problems, we propose a new non-autoregressive transformer with a unified bidirectional decoder (NAT-UBD), which can simultaneously utilize left-to-right and right-to-left contexts. However, direct use of bidirectional contexts will cause information leakage, which means the decoder output can be affected by the character information from the input of the same position. To avoid information leakage, we propose a novel attention mask and modify vanilla queries, keys, and values matrices for NAT-UBD. Experimental results verify that NAT-UBD can achieve character error rates (CERs) of 5.0%/5.5% on the Aishell1 dev/test sets, outperforming all previous NAR transformer models. Moreover, NAT-UBD can run 49.8x faster than the AR transformer baseline when decoding in a single step.

cs.CL↗

arXiv2021

Ultrahigh-Efficiency Second Harmonic Generation in a Reverse-Polarization Dual-Layer Crystalline Thin Film Nanophotonic Waveguide

Second harmonic generation (SHG), as one of the most significant \c{hi}(2) nonlinear optical processes, plays crucial roles in a broad variety of optical and photonic applications. Designing various delicate schemes to achieve highly efficient SHG has become a long standing and challenging topic in field of nonlinear optics. Despite numerous success on SHG based on birefringent phase matching and quasi-phase matching, so far, modal phase matching (MPM) for SHG in tightly light-confined structures has still in its infancy. Here, we propose a new scheme to realize highly-efficient SHG via MPM by using a nanophotonic LiNbO3 thin-film waveguide consists of two bonded layers with internally reversed polarizations. In such a dual-layer ridge waveguide based on lithium niobate on insulator, upon optical excitation at 1574.6 nm, we observe SHG at 787.3 nm with ultrahigh conversion efficiency of 5,540% /W/cm/cm experimentally. This work advances our understanding on modal-phase-matched SHG and other quadratic optical nonlinear process, offering additional strategies for development of high-performance nonlinear photonic devices in on-chip platforms.

physics.optics↗

arXiv2021

Fairness-Aware Online Meta-learning

In contrast to offline working fashions, two research paradigms are devised for online learning: (1) Online Meta Learning (OML) learns good priors over model parameters (or learning to learn) in a sequential setting where tasks are revealed one after another. Although it provides a sub-linear regret bound, such techniques completely ignore the importance of learning with fairness which is a significant hallmark of human intelligence. (2) Online Fairness-Aware Learning. This setting captures many classification problems for which fairness is a concern. But it aims to attain zero-shot generalization without any task-specific adaptation. This therefore limits the capability of a model to adapt onto newly arrived data. To overcome such issues and bridge the gap, in this paper for the first time we proposed a novel online meta-learning algorithm, namely FFML, which is under the setting of unfairness prevention. The key part of FFML is to learn good priors of an online fair classification model's primal and dual parameters that are associated with the model's accuracy and fairness, respectively. The problem is formulated in the form of a bi-level convex-concave optimization. Theoretic analysis provides sub-linear upper bounds for loss regret and for violation of cumulative fairness constraints. Our experiments demonstrate the versatility of FFML by applying it to classification on three real-world datasets and show substantial improvements over the best prior work on the tradeoff between fairness and classification accuracy

cs.LG↗

arXiv2021

Network-wide link travel time and station waiting time estimation using automatic fare collection data: A computational graph approach

Urban rail transit (URT) system plays a dominating role in many megacities like Beijing and Hong Kong. Due to its important role and complex nature, it is always in great need for public agencies to better understand the performance of the URT system. This paper focuses on an essential and hard problem to estimate the network-wide link travel time and station waiting time using the automatic fare collection (AFC) data in the URT system, which is beneficial to better understand the system-wide real-time operation state. The emerging data-driven techniques, such as computational graph (CG) models in the machine learning field, provide a new solution for solving this problem. In this study, we first formulate a data-driven estimation optimization framework to estimate the link travel time and station waiting time. Then, we cast the estimation optimization model into a CG framework to solve the optimization problem and obtain the estimation results. The methodology is verified on a synthetic URT network and applied to a real-world URT network using the synthetic and real-world AFC data, respectively. Results show the robustness and effectiveness of the CG-based framework. To the best of our knowledge, this is the first time that the CG is applied to the URT. This study can provide critical insights to better understand the operational state in URT.

cs.LG↗

arXiv2021

Deep-learning Architecture for Short-term Passenger Flow Forecasting in Urban Rail Transit

Short-term passenger flow forecasting is an essential component in urban rail transit operation. Emerging deep learning models provide good insight into improving prediction precision. Therefore, we propose a deep learning architecture combining the residual network (ResNet), graph convolutional network (GCN), and long short-term memory (LSTM) (called "ResLSTM") to forecast short-term passenger flow in urban rail transit on a network scale. First, improved methodologies of the ResNet, GCN, and attention LSTM models are presented. Then, the model architecture is proposed, wherein ResNet is used to capture deep abstract spatial correlations between subway stations, GCN is applied to extract network topology information, and attention LSTM is used to extract temporal correlations. The model architecture includes four branches for inflow, outflow, graph-network topology, as well as weather conditions and air quality. To the best of our knowledge, this is the first time that air-quality indicators have been taken into account, and their influences on prediction precision quantified. Finally, ResLSTM is applied to the Beijing subway using three time granularities (10, 15, and 30 min) to conduct short-term passenger flow forecasting. A comparison of the prediction performance of ResLSTM with those of many state-of-the-art models illustrates the advantages and robustness of ResLSTM. Moreover, a comparison of the prediction precisions obtained for time granularities of 10, 15, and 30 min indicates that prediction precision increases with increasing time granularity. This study can provide subway operators with insight into short-term passenger flow forecasting by leveraging deep learning models.

eess.SP↗

arXiv2021

Adaptive Multi-modal Fusion Hashing via Hadamard Matrix

Hashing plays an important role in information retrieval, due to its low storage and high speed of processing. Among the techniques available in the literature, multi-modal hashing, which can encode heterogeneous multi-modal features into compact hash codes, has received particular attention. Most of the existing multi-modal hashing methods adopt the fixed weighting factors to fuse multiple modalities for any query data, which cannot capture the variation of different queries. Besides, many methods introduce hyper-parameters to balance many regularization terms that make the optimization harder. Meanwhile, it is time-consuming and labor-intensive to set proper parameter values. The limitations may significantly hinder their promotion in real applications. In this paper, we propose a simple, yet effective method that is inspired by the Hadamard matrix. The proposed method captures the multi-modal feature information in an adaptive manner and preserves the discriminative semantic information in the hash codes. Our framework is flexible and involves a very few hyper-parameters. Extensive experimental results show the method is effective and achieves superior performance compared to state-of-the-art algorithms.

cs.MM↗

arXiv2021

Forward Modeling of Solar Coronal Magnetic Field Measurements Based on a Magnetic-field-induced Transition in Fe X

It was recently proposed that the intensity ratios of several extreme ultraviolet spectral lines from the Fe X ion can be used to measure the solar coronal magnetic field based on the magnetic-field-inducedtransition (MIT) theory. To verify the suitability of this method, we performed forward modelingwith a three-dimensional radiation magnetohydrodynamic model of a solar active region. Intensities of several spectral lines from Fe X were synthesized from the model. Based on the MIT theory, intensity ratios of the MIT line Fe X 257 A to several other Fe X lines were used to derive the magnetic field strengths, which were then compared with the field strengths in the model. We also developed a new method to simultaneously estimate the coronal density and temperature from the Fe X 174/175 and 184/345 A line ratios. Using these estimates, we demonstrated that the MIT technique can provide reasonably accurate measurements of the coronal magnetic field in both on-disk and off-limb solar observations. Our investigation suggests that a spectrometer that can simultaneously observe the Fe X 174, 175, 184, 257, and 345 A lines and allow an accurate radiometric calibration for these lines is highly desired to achieve reliable measurements of the coronal magnetic field. We have also evaluatedthe impact of the uncertainty in the Fe X 3p4 3d 4D5/2 and 4D7/2 energy difference on the magnetic field measurements.

astro-ph.SR↗

arXiv2021

Bridging the Gap between Spatial and Spectral Domains: A Survey on Graph Neural Networks

Deep learning's success has been widely recognized in a variety of machine learning tasks, including image classification, audio recognition, and natural language processing. As an extension of deep learning beyond these domains, graph neural networks (GNNs) are designed to handle the non-Euclidean graph-structure which is intractable to previous deep learning techniques. Existing GNNs are presented using various techniques, making direct comparison and cross-reference more complex. Although existing studies categorize GNNs into spatial-based and spectral-based techniques, there hasn't been a thorough examination of their relationship. To close this gap, this study presents a single framework that systematically incorporates most GNNs. We organize existing GNNs into spatial and spectral domains, as well as expose the connections within each domain. A review of spectral graph theory and approximation theory builds a strong relationship across the spatial and spectral domains in further investigation.

cs.LG↗

arXiv2021

CRIL: Continual Robot Imitation Learning via Generative and Prediction Model

Imitation learning (IL) algorithms have shown promising results for robots to learn skills from expert demonstrations. However, they need multi-task demonstrations to be provided at once for acquiring diverse skills, which is difficult in real world. In this work we study how to realize continual imitation learning ability that empowers robots to continually learn new tasks one by one, thus reducing the burden of multi-task IL and accelerating the process of new task learning at the same time. We propose a novel trajectory generation model that employs both a generative adversarial network and a dynamics-aware prediction model to generate pseudo trajectories from all learned tasks in the new task learning process. Our experiments on both simulation and real-world manipulation tasks demonstrate the effectiveness of our method.

cs.RO↗

arXiv2021

Brain-inspired global-local learning incorporated with neuromorphic computing

Two main routes of learning methods exist at present including error-driven global learning and neuroscience-oriented local learning. Integrating them into one network may provide complementary learning capabilities for versatile learning scenarios. At the same time, neuromorphic computing holds great promise, but still needs plenty of useful algorithms and algorithm-hardware co-designs for exploiting the advantages. Here, we report a neuromorphic hybrid learning model by introducing a brain-inspired meta-learning paradigm and a differentiable spiking model incorporating neuronal dynamics and synaptic plasticity. It can meta-learn local plasticity and receive top-down supervision information for multiscale synergic learning. We demonstrate the advantages of this model in multiple different tasks, including few-shot learning, continual learning, and fault-tolerance learning in neuromorphic vision sensors. It achieves significantly higher performance than single-learning methods, and shows promise in empowering neuromorphic applications revolution. We further implemented the hybrid model in the Tianjic neuromorphic platform by exploiting algorithm-hardware co-designs and proved that the model can fully utilize neuromorphic many-core architecture to develop hybrid computation paradigm.

cs.NE↗

arXiv2021

FDNet: A Deep Learning Approach with Two Parallel Cross Encoding Pathways for Precipitation Nowcasting

With the goal of predicting the future rainfall intensity in a local region over a relatively short period time, precipitation nowcasting has been a long-time scientific challenge with great social and economic impact. The radar echo extrapolation approaches for precipitation nowcasting take radar echo images as input, aiming to generate future radar echo images by learning from the historical images. To effectively handle complex and high non-stationary evolution of radar echoes, we propose to decompose the movement into optical flow field motion and morphologic deformation. Following this idea, we introduce Flow-Deformation Network (FDNet), a neural network that models flow and deformation in two parallel cross pathways. The flow encoder captures the optical flow field motion between consecutive images and the deformation encoder distinguishes the change of shape from the translational motion of radar echoes. We evaluate the proposed network architecture on two real-world radar echo datasets. Our model achieves state-of-the-art prediction results compared with recent approaches. To the best of our knowledge, this is the first network architecture with flow and deformation separation to model the evolution of radar echoes for precipitation nowcasting. We believe that the general idea of this work could not only inspire much more effective approaches but also be applied to other similar spatiotemporal prediction tasks

cs.LG↗

arXiv2021

Multidimensional Uncertainty-Aware Evidential Neural Networks

Traditional deep neural networks (NNs) have significantly contributed to the state-of-the-art performance in the task of classification under various application domains. However, NNs have not considered inherent uncertainty in data associated with the class probabilities where misclassification under uncertainty may easily introduce high risk in decision making in real-world contexts (e.g., misclassification of objects in roads leads to serious accidents). Unlike Bayesian NN that indirectly infer uncertainty through weight uncertainties, evidential NNs (ENNs) have been recently proposed to explicitly model the uncertainty of class probabilities and use them for classification tasks. An ENN offers the formulation of the predictions of NNs as subjective opinions and learns the function by collecting an amount of evidence that can form the subjective opinions by a deterministic NN from data. However, the ENN is trained as a black box without explicitly considering inherent uncertainty in data with their different root causes, such as vacuity (i.e., uncertainty due to a lack of evidence) or dissonance (i.e., uncertainty due to conflicting evidence). By considering the multidimensional uncertainty, we proposed a novel uncertainty-aware evidential NN called WGAN-ENN (WENN) for solving an out-of-distribution (OOD) detection problem. We took a hybrid approach that combines Wasserstein Generative Adversarial Network (WGAN) with ENNs to jointly train a model with prior knowledge of a certain class, which has high vacuity for OOD samples. Via extensive empirical experiments based on both synthetic and real-world datasets, we demonstrated that the estimation of uncertainty by WENN can significantly help distinguish OOD samples from boundary samples. WENN outperformed in OOD detection when compared with other competitive counterparts.

cs.LG↗

arXiv2021

Generating Adjacency-Constrained Subgoals in Hierarchical Reinforcement Learning

Goal-conditioned hierarchical reinforcement learning (HRL) is a promising approach for scaling up reinforcement learning (RL) techniques. However, it often suffers from training inefficiency as the action space of the high-level, i.e., the goal space, is often large. Searching in a large goal space poses difficulties for both high-level subgoal generation and low-level policy learning. In this paper, we show that this problem can be effectively alleviated by restricting the high-level action space from the whole goal space to a $k$-step adjacent region of the current state using an adjacency constraint. We theoretically prove that the proposed adjacency constraint preserves the optimal hierarchical policy in deterministic MDPs, and show that this constraint can be practically implemented by training an adjacency network that can discriminate between adjacent and non-adjacent subgoals. Experimental results on discrete and continuous control tasks show that incorporating the adjacency constraint improves the performance of state-of-the-art HRL approaches in both deterministic and stochastic environments.

cs.LG↗

arXiv2021

Exploring the Nature of EUV Waves in a Radiative Magnetohydrodynamic Simulation

Coronal extreme-ultraviolet (EUV) waves are large-scale disturbances propagating in the corona, whose physical nature and origin have been discussed for decades. We report the first three dimensional (3D) radiative magneto-hydrodynamic (RMHD) simulation of a coronal EUV wave and the accompanying quasi-periodic wave trains. The numerical experiment is conducted with the MURaM code and simulates the formation of solar active regions through magnetic flux emergence from the convection zone to the corona. The coronal EUV wave is driven by the eruption of a magnetic flux rope that also gives rise to a C-class flare. It propagates in a semi-circular shape with an initial speed ranging from about 550 to 700 km s$^{-1}$, which corresponds to an average Mach number (relative to fast magnetoacoustic waves) of about 1.2. Furthermore, the abrupt increase of the plasma density, pressure and tangential magnetic field at the wavefront confirms fast-mode shock nature of the coronal EUV wave. Quasi-periodic wave trains with a period of about 30 s are found as multiple secondary wavefronts propagating behind the leading wavefront and ahead of the erupting magnetic flux rope. We also note that the true wavefront in the 3D space can be very inhomogeneous, however, the line-of-sight integration of EUV emission significantly smoothes the sharp structures in 3D and leads to a more diffuse wavefront.

astro-ph.SR↗

arXiv2021

A Primal-Dual Subgradient Approachfor Fair Meta Learning

The problem of learning to generalize to unseen classes during training, known as few-shot classification, has attracted considerable attention. Initialization based methods, such as the gradient-based model agnostic meta-learning (MAML), tackle the few-shot learning problem by "learning to fine-tune". The goal of these approaches is to learn proper model initialization, so that the classifiers for new classes can be learned from a few labeled examples with a small number of gradient update steps. Few shot meta-learning is well-known with its fast-adapted capability and accuracy generalization onto unseen tasks. Learning fairly with unbiased outcomes is another significant hallmark of human intelligence, which is rarely touched in few-shot meta-learning. In this work, we propose a Primal-Dual Fair Meta-learning framework, namely PDFM, which learns to train fair machine learning models using only a few examples based on data from related tasks. The key idea is to learn a good initialization of a fair model's primal and dual parameters so that it can adapt to a new fair learning task via a few gradient update steps. Instead of manually tuning the dual parameters as hyperparameters via a grid search, PDFM optimizes the initialization of the primal and dual parameters jointly for fair meta-learning via a subgradient primal-dual approach. We further instantiate examples of bias controlling using mean difference and decision boundary covariance as fairness constraints to each task for supervised regression and classification, respectively. We demonstrate the versatility of our proposed approach by applying our approach to various real-world datasets. Our experiments show substantial improvements over the best prior work for this setting.

cs.LG↗

arXiv2021

Faster State Preparation across Quantum Phase Transition Assisted by Reinforcement Learning

An energy gap develops near quantum critical point of quantum phase transition in a finite many-body (MB) system, facilitating the ground state transformation by adiabatic parameter change. In real application scenarios, however, the efficacy for such a protocol is compromised by the need to balance finite system life time with adiabaticity, as exemplified in a recent experiment that prepares three-mode balanced Dicke state near deterministically [PNAS {\bf 115}, 6381 (2018)]. Instead of tracking the instantaneous ground state as unanimously required for most adiabatic crossing, this work reports a faster sweeping policy taking advantage of excited level dynamics. It is obtained based on deep reinforcement learning (DRL) from a multi-step training scheme we develop. In the absence of loss, a fidelity $\ge 99\%$ between prepared and the target Dicke state is achieved over a small fraction of the adiabatically required time. When loss is included, training is carried out according to an operational benchmark, the interferometric sensitivity of the prepared state instead of fidelity, leading to better sensitivity in about half of the previously reported time. Implemented in a Bose-Einstein condensate of $\sim 10^4$ $^{87}$Rb atoms, the balanced three-mode Dicke state exhibiting an improved number squeezing of $13.02\pm0.20$ dB is observed within 766 ms, highlighting the potential of DRL for quantum dynamics control and quantum state preparation in interacting MB systems.

cond-mat.quant-gas↗